Comment by orbital-decay
1 day ago
>Dynamically adjust the difficulty depending on how well the player is doing. <- This is actually magic sauce and could be the topic of a whole talk.
Please never do this. This deceptively looks smart to a game designer, but actual players read this like a book and then it feels cheap and massively unrewarding for them.
This depends a lot on the game and the situation. If hidden well enough i suspect it can play well. I really enjoyed the difficulty of Half Life Alyx for instance. I was saving my resources carefully because they seemed very sparse.
But then I noticed that the loot was entirely dynamic. I forget the exact items, but if you only use one grenade it will only generate one refill hidden around the next area. If you use all your resources though, the upcoming loot gets massively increased and you can usually get almost all of it back right away. I learned this partway through and completely changed my playstyle and really broke the immersion to see the game difficulty change in realtime. I reloaded a save and played differently, and the next area was filled with loot instead of empty. But when asking my friends what they thought, none of them noticed that the game was doing this, and they all thought it was well balanced.
On the other hand, some Mario games now will essentially have a "you are dying a lot, here is a star power up to help you out, or heres luigi to play the level for you" mechanic. For a super casual game, such a mechanic is great. Its optional. If you want to lock-in and beat the hard level, you can just not grab the powerup, or say no to the prompt. But if you are getting frustrated, you can skip the hard parts and fly through to the next level and get right back to having fun.
The Alyx example is interesting, because I see that system as being designed to avoid breaking immersion. Specifically, to avoid entering an "unwinnable" state i.e. where the player restarts from a save/checkpoint. I do understand how noticing the underlying system can pull you out of the game though.
Intuitively, people want to save up ammo because they never know when there's gonna be a big fight, and they don't want to be caught low on ammo.
I had the same experience, I was trying to be very accurate with the guns and thought, wow, there's hardly any ammo to pick up in this game. Partway through I got tired of being stingy and started being less careful, having fun going rambo, and suddenly ammo was spawning everywhere.
I think it would be much better if the game just spawned enough to max out your pools fairly early, so you can focus less on inventory and more on combat. Then they can adjust the size of the ammo pools to tune how accurate they want you to be.
I think part of the issue is that Alyx doesn't really have any other options to fight besides guns that need ammo or however many grenades you're choosing to carry. They didn't have melee as a design decision, but it means they need to keep you topped up on ammo or you're defenseless.
> But when asking my friends what they thought, none of them noticed that the game was doing this, and they all thought it was well balanced.
Sounds like the design is working as intended?
The issue IMO is that there's a subset of gamers who really do not want games to try to balance things automatically, e.g. people who really hate skill-based matchmaking. Most people will enjoy the system, but this subset sees a balancing feedback loop as punishment for being "too good".
Agreed. Something like Dark Souls where an enemy is an enemy and it'll smash you if you're underleveled is so much more satisfying than something like Oblivion where enemy health inexplicably goes up based on your level. It also gives you the satisfaction of going back to earlier areas that were scary and laughing off the puny enemies.
Totally different when it comes to multiplayer matchmaking, though. There I wish games were even more aggressive and exact at matching you to players around your skill level. Nothing worse than casually sitting down after work for a quick multiplayer PVP game of any sort and getting totally annihilated by a streamer who plays 20 hours per day.
I used to really enjoy this aspect of older games like Warcraft 3 and TF2.
You'd get the one guy on your server who was just in another league compared to everyone else, and it was so satisfying to get the kill on him. And sometimes, it would be you pubstomping.
Then again, back then I don't think people cared as much about winning and losing.
1 reply →
There's a trade-off between how long you wait and the skill spread of the team. IIRC at the very top, for DOTA2, the pros complained enough for the devs to make a "wait longer" matchmaking mode.
> There I wish games were even more aggressive and exact at matching you to players around your skill level.
PvP involves sandbagging and a variety of fragile gimmick strategies. (Oh, and cheating. For slow enough turn-based games, players can cheat with a separate physical device, which can't be detected.)
The system can't measure skill level accurately against a player base that's actively antagonistic to such measurement, so of course it can't match players well on that basis.
Even abstract games of perfect information, like chess and Go, have this problem online.
SBMM also a double-edged sword. Some people like to play game relaxed. Random matchmaking gives them chance to win. But SBMM means having sweaty games all the time. I do prefer SBMM but I’ve seen people upset about it
5 replies →
It doesn't have to be deceptive!
In Factorio, you build an automated factory on an alien world whose native inhabitants hate air pollution.
If you spend your time gathering resources manually or with a humble-sized factory, there'd be very little alien hostility. But the better you're playing, the grander the factories, the more you're polluting, the more aggressive the aliens attack you.
Furthermore, trees soak up CO2, so if you preserve them you reduce alien hostility, but if you go fully industrious and build on top of them you invite more alien aggression.
I think this concept is delightful and far from deceptive. It makes the game suitable for players of all levels and it meshes with the game mechanism rather than working behind your back.
hnews folks would also probably enjoy the cellular-automata-like mechanism of the pollution. Essentially the world is divided into cells and each cell has a pollution value. Pollution dissipates to nearby cells on regular intervals. Cells with trees soak up some pollution, cells with machines produce some pollution. A pollution map is available in-game.
A big advantage of this system is the transparency. You understand the logic behind it so it doesn't feel like someone is trying to trick you.
IIRC, the playtime is also a factor in Factorio. It's nice that the trees soak up pollution, but the inability to plant more of them is actually the only thing I really dislike about the game. I want to have a choice to be even greener and even less disruptive to the local inhabitants in the base game, not as a mod "hack"
You can plant trees now.
> then it feels cheap and massively unrewarding for them
Or even blocks them, in games that have dynamic leveling. Oblivion is the ur-example, but I've ran across at least two more:
1. Been playing Tyranny and there are mentions in the faqs/forums that you should disable leveling of skills you don't use if you do the new game+ thing where there are no level caps.
2. And one that happened to me personally a year or two ago: was playing Grim Dawn and hit my character's incompetence ceiling. This is a gear based ARPG/Diablo clone so the solution is to get better gear. But you can't go back to an easier area to improve your gear because they're all auto scaled and every map is the same - too hard now - difficulty. So there goes the uninstall and the lost DLC sales...
Hard disagree. All of my favourite games of all time have dynamic difficulty systems.
Not stupid stuff like levelling up enemies in Bethesda games - but actual skill based games that get harder depending on how good you are at it: Resident Evil 4, Battle Garegga to name a few.
Games that unlock more content as you reach skill gates are the most memorable.
It's often a very bad idea. One example is The Elder Scrolls IV: Oblivion. Players are punished for leveling, and many players simply give up and have to do everything on level 1.
It's essentially punishing players for playing well, and often could be gamed by playing badly intentionally.
Oblivion's problem is even worse than that. You can level up entirely from improving non-combat skills, so the enemies are harder and you're only better at jumping.
I think it strongly depends on how it's done and how it plays into the feeling of the game. Certain kinds of level scaling can feel pretty cheap, for example, because they kind of work against the feeling of progression you get in the game. Similarly a game obviously making things easier for you can feel like it's cheapening the satisfaction of improving at the game, though how important this is varies from person to person.
But done well, it can much more reliably create the feeling of being just on the edge of failure but pulling through, which is really satisfying in a lot of games. L4D did this, but Helldivers 2 for example also has a lot of game design focused on creating this kind of feeling.
On the other hand, the whole Director system of Left 4 Dead works like this and it’s incredible.
https://cdn.akamai.steamstatic.com/apps/valve/2009/ai_system...
It's a pretty well proven design pattern.
The adage is that the customers never know what they want, and it's doubly true when it comes to video games.
My go-to example is Homeworld 2, where the dynamic difficulty scaling was IMO very poorly hidden. It was as if the player is being punished for doing the legwork to preserve and expand their fleet across missions even though the game's lore and mechanics encourage them to do so.
Crash Bandicoot did this well. In the original game, you needed to finish without dying and get all of the boxes to get the gem. In future games, some levels would have a “death path” that would only be available if you got to a certain point in the level without dying.
Yes feels like this can easily be "gamed"
For example, if I want friendlier lobbies in arc raiders, I will spawn in with no loadout and purposely surrender.. I can game my difficulty lower
Ooh very much disagree. Like so many things, when it’s done right, it works real well, it feels good. See Deathloop for a recent example.
The good way of doing this is to ask the player if the difficulty shall be lowered temporarily just for that level of they keep dying. Then it is transparent and the choice is theirs.
"actual players" as opposed to other people who just, what? "use" the game?
Sorry, but it's hard to take this comment seriously when it reeks of "git gud". So I'll just point out that a lot of people play games, and that there are well-documented instances of games with dynamic difficulty as early as the 80s.
As opposed to the game designer who came up with an idea that sounds convincing but barely ever tested it. My comment had nothing to do with "git gud", it's poorly worded perhaps.
Of course there's a ton of this, which is exactly why I complain. Level scaling is probably the easiest to justify because it solves legitimate issues (that are often caused by other poor decisions), but other types are usually much worse.
> My comment had nothing to do with "git gud"
Fair enough, I applied that lens to it on my own.
Allow me to elaborate on my earlier comment and reply to this one here:
A ton of it? No, I don't see that. It's probably the least commonly implemented approach to difficulty (curves).
Dynamically adjusted difficulty has been around for a long time, but it's never been broadly adopted. There's a lot of reasons for this I'm sure, but a couple of obvious ones are that executives don't tend to throw money at features they can't see [or game developers in general], and that it's a tricky thing to build anyway.
It's certainly not the default solution to the problem, but there's a long history of games implementing some form of dynamically adjusted difficulty. Some of them to great success with the difficulty system itself being celebrated (Xevious, Resident Evil 4, Left 4 Dead), the vast majority of them you have no idea even implemented such a feature, and there's also a number of famous examples that were widely derided for it (Oblivion, ...?). [I'm genuinely stumped to think of another famously bad example, not trying to make it look like there are fewer of one than the other. I'm sure there are just as many negatively viewed ones as I can come up with positive ones.]
While it's obviously possible that an individual designer/team didn't thoroughly test their particular feature, the concept itself has been very well tested in a wide variety of implementations/mechanics. And if you're playing a game designed by people who didn't test it, the difficulty scaling seems likely to be the least of your problems.
Left 4 Dead handled this pretty well in my opinion (i.e. mob spawning behind if you're playing too slowly or a tank in front if you're going too quick). I absolutely hated the dynamic adjusting in games like Mario Kart, though.
in games like wow where the more powerful you get the harder the enemies become, this reduces incentive to get more powerful. its a simple consequence so im not sure why they do it - probably to keep "time played" metric hovering at the desired level. which imho backfires when time played becomes 0.
You're only looking at one side. Dynamic difficulty also allows players to progress by lowering the health of a boss they've failed to beat for over two hours.
Casual players giving up was a problem that no amount of shaming managed to fix ("haha skill issue, sucks to suck").
How is it a problem? Giving up is absolutely fine. Different games are made for different types of players. Who cares what the others think (or you think they think). By trying to cater to everyone you satisfy nobody.
2 replies →
Resident Evil 4 famously had a hidden dynamic difficulty setting that made the game easier if the player struggled. I think it worked because RE4 is a linear / ambiance based game. The game designer wants the player to progress in the story, and getting stuck on an impossible doesn't make sense except if the player chose a challenge mode. Contrast with Morrowing where it makes no sense having common bandit on the road carrying god-like equipment.
Just make it a concrete option then. What is the point of being sneaky about it?
2 replies →
This is why Diablo 4 was a complete flop.
> This is why Diablo 4 was a complete flop.
This is interesting to me; can you explain please?
Difficulty scaling is done to such an extent that even common enemies scale right along with you. In the opening scene, there is a bear you come across and at level 1 or 2, you are able to one shot it. Then the story takes you on a grand quest where ostensibly you go from a nobody to the savior of the world. However, due to difficulty scaling, if you run into the same bear halfway through the game, it will be harder to kill than when you were level 1.
I like when your character or your party feels like superheroes for a little while after leveling up, then it gets harder to really test your new ability, then keep repeating that cycle.
Do you really hate Breathe of the Wild so much?